This invention relates to a novel method of assembling a CRT (cathode-ray tube) using a coded subassembly or part.
In manufacturing a color television picture tube, which is a CRT, it is the practice to assemble a succession of different subassemblies to one particular part. In one procedure, for example, the mask-frame subassembly is assembled to the panel, then the viewing screen subassembly is assembled to the panel, then the funnel assembly is assembled to the panel, then the electron-gun mount subassembly is assembled to the panel and then, after exhausting, sealing and electrode processing, the implosion-protection subassembly is assembled to the panel. Subsequently, the tube is tested, packed, warehoused, and distributed. The procedure may be more complex if, as occurs in some factories, different sizes and types of product are intermixed during one or more of the steps.
Attempts have been made toward conducting these assembling steps by machine, as well as making each of the subassemblies by machine. These attempts have all employed some amount of human recognition by sight and/or touch. Where machine recognition has been used, the different techniques used for different subassemblies have not been interrelated and in many cases have been incompatible. Furthermore, few of the prior techniques have provided historical data which could be used, for example, for tracing the causes of out-of-spectification product, for warehousing and cost analyses, or for warranty purposes.
However, as the product line becomes more mature, profit margins erode, and there is a need to reduce all product costs to a minimum. This can be done by reducing assembly costs and by the earliest detection and correction of defects. This can be achieved with full automation provided the cost of full automation is not greater than the cost of the prior systems it replaces. One feature of practical full automation is that it can be phased into and out of existing manual and partially-automated portions of the total system.